本文转自:http://poi.apache.org/spreadsheet/user-defined-functions.html

How to Create and Use User Defined Functions

Description

This document describes the User Defined Functions within POI.            User defined functions allow you to take code that is written in VBA            and re-write in Java and use within POI. Consider the following example.

An Example

Suppose you are given a spreadsheet that can calculate the principal and interest        payments for a mortgage.  The user enters the principal loan amount, the interest rate        and the term of the loan.  The Excel spreadsheet does the rest.

When you actually look at the workbook you discover that rather than having        the formula in a cell it has been written as VBA function.  You review the          function and determine that it could be written in Java:

If we write a small program to try to evaluate this cell, we'll fail.  Consider this source code:

import java.io.File ;
import java.io.FileInputStream ;
import java.io.FileNotFoundException ;
import java.io.IOException ; import org.apache.poi.openxml4j.exceptions.InvalidFormatException ;
import org.apache.poi.ss.formula.functions.FreeRefFunction ;
import org.apache.poi.ss.formula.udf.AggregatingUDFFinder ;
import org.apache.poi.ss.formula.udf.DefaultUDFFinder ;
import org.apache.poi.ss.formula.udf.UDFFinder ;
import org.apache.poi.ss.usermodel.Cell ;
import org.apache.poi.ss.usermodel.CellValue ;
import org.apache.poi.ss.usermodel.Row ;
import org.apache.poi.ss.usermodel.Sheet ;
import org.apache.poi.ss.usermodel.Workbook ;
import org.apache.poi.ss.usermodel.WorkbookFactory ;
import org.apache.poi.ss.util.CellReference ; public class Evaluator { public static void main( String[] args ) { System.out.println( "fileName: " + args[0] ) ;
System.out.println( "cell: " + args[1] ) ; File workbookFile = new File( args[0] ) ; try {
FileInputStream fis = new FileInputStream(workbookFile);
Workbook workbook = WorkbookFactory.create(fis); FormulaEvaluator evaluator = workbook.getCreationHelper().createFormulaEvaluator(); CellReference cr = new CellReference( args[1] ) ;
String sheetName = cr.getSheetName() ;
Sheet sheet = workbook.getSheet( sheetName ) ;
int rowIdx = cr.getRow() ;
int colIdx = cr.getCol() ;
Row row = sheet.getRow( rowIdx ) ;
Cell cell = row.getCell( colIdx ) ; CellValue value = evaluator.evaluate( cell ) ; System.out.println("returns value: " + value ) ; } catch( FileNotFoundException e ) {
e.printStackTrace();
} catch( InvalidFormatException e ) {
e.printStackTrace();
} catch( IOException e ) {
e.printStackTrace();
}
}
}

If you run this code, you're likely to get the following error:

Exception in thread "main" org.apache.poi.ss.formula.eval.NotImplementedException: Error evaluating cell Sheet1!B4
at org.apache.poi.ss.formula.WorkbookEvaluator.addExceptionInfo(WorkbookEvaluator.java:321)
at org.apache.poi.ss.formula.WorkbookEvaluator.evaluateAny(WorkbookEvaluator.java:288)
at org.apache.poi.ss.formula.WorkbookEvaluator.evaluate(WorkbookEvaluator.java:221)
at org.apache.poi.hssf.usermodel.HSSFFormulaEvaluator.evaluateFormulaCellValue(HSSFFormulaEvaluator.java:320)
at org.apache.poi.hssf.usermodel.HSSFFormulaEvaluator.evaluate(HSSFFormulaEvaluator.java:182)
at poi.tests.Evaluator.main(Evaluator.java:61)
Caused by: org.apache.poi.ss.formula.eval.NotImplementedException: calculatePayment
at org.apache.poi.ss.formula.UserDefinedFunction.evaluate(UserDefinedFunction.java:59)
at org.apache.poi.ss.formula.OperationEvaluatorFactory.evaluate(OperationEvaluatorFactory.java:129)
at org.apache.poi.ss.formula.WorkbookEvaluator.evaluateFormula(WorkbookEvaluator.java:456)
at org.apache.poi.ss.formula.WorkbookEvaluator.evaluateAny(WorkbookEvaluator.java:279)
... 4 more

How would we make it so POI can use this sheet?

Defining Your Function

To 'convert' this code to Java and make it available to POI you need to implement        a FreeRefFunction instance.  FreeRefFunction is an interface in the org.apache.poi.ss.formula.functions         package.  This interface defines one method, evaluate(ValueEval[] args, OperationEvaluationContext ec),        which is how you will receive the argument values from POI.

The evaluate() method as defined above is where you will convert the ValueEval instances to the         proper number types.  The following code snippet shows you how to get your values:

public class CalculateMortgage implements FreeRefFunction {

@Override
public ValueEval evaluate( ValueEval[] args, OperationEvaluationContext ec ) {
if (args.length != 3) {
return ErrorEval.VALUE_INVALID;
} double principal, rate, years, result;
try {
ValueEval v1 = OperandResolver.getSingleValue( args[0],
ec.getRowIndex(),
ec.getColumnIndex() ) ;
ValueEval v2 = OperandResolver.getSingleValue( args[1],
ec.getRowIndex(),
ec.getColumnIndex() ) ;
ValueEval v3 = OperandResolver.getSingleValue( args[2],
ec.getRowIndex(),
ec.getColumnIndex() ) ; principal = OperandResolver.coerceValueToDouble( v1 ) ;
rate = OperandResolver.coerceValueToDouble( v2 ) ;
years = OperandResolver.coerceValueToDouble( v3 ) ;

The first thing we do is check the number of arguments being passed since there is no sense     in attempting to go further if you are missing critical information.

Next we declare our variables, in our case we need variables for:

  • principal - the amount of the loan
  • rate - the interest rate as a decimal
  • years - the length of the loan in years
  • result - the result of the calculation

Next, we use the OperandResolver to convert the ValueEval instances to doubles, though not directly.       First we start by getting discreet values.  Using the OperandResolver.getSingleValue() method     we retrieve each of the values passed in by the cell in the spreadsheet.  Next, we use the     OperandResolver again to convert the ValueEval instances to doubles, in this case.  This     class has other methods of coercion for gettings Strings, ints and booleans.  Now that we've      got our primitive values we can move on to calculating the value.

As shown previously, we have the VBA source.  We need to add code to our class to calculate      the payment.  To do this you could simply add it to the method we've already created but I've     chosen to add it as its own method.  Add the following method:

public double calculateMortgagePayment( double p, double r, double y ) {

    double i = r / 12 ;
double n = y * 12 ; double principalAndInterest =
p * (( i * Math.pow((1 + i),n ) ) / ( Math.pow((1 + i),n) - 1)) ; return principalAndInterest ;
}

The biggest change necessary is related to the exponents; Java doesn't have a notation for this     so we had to add calls to Math.pow().  Now we need to add this call to our previous method:

     	 result = calculateMortgagePayment( principal, rate, years ) ;
     		

Having done that, the last things we need to do are to check to make sure we didn't get a bad result and,     if not, we need to return the value. Add the following code to the class:

private void checkValue(double result) throws EvaluationException {
if (Double.isNaN(result) || Double.isInfinite(result)) {
throw new EvaluationException(ErrorEval.NUM_ERROR);
}
}

Then add a line of code to our evaluate method to call this new static method, complete our try/catch and return the value:

        checkValue(result);

    } catch (EvaluationException e) {
e.printStackTrace() ;
return e.getErrorEval();
} return new NumberEval( result ) ;

So the whole class would be as follows:

import org.apache.poi.ss.formula.OperationEvaluationContext ;
import org.apache.poi.ss.formula.eval.ErrorEval ;
import org.apache.poi.ss.formula.eval.EvaluationException ;
import org.apache.poi.ss.formula.eval.NumberEval ;
import org.apache.poi.ss.formula.eval.OperandResolver ;
import org.apache.poi.ss.formula.eval.ValueEval ;
import org.apache.poi.ss.formula.functions.FreeRefFunction ; /**
* A simple function to calculate principal and interest.
*
* @author Jon Svede
*
*/
public class CalculateMortgage implements FreeRefFunction { @Override
public ValueEval evaluate( ValueEval[] args, OperationEvaluationContext ec ) {
if (args.length != 3) {
return ErrorEval.VALUE_INVALID;
} double principal, rate, years, result;
try {
ValueEval v1 = OperandResolver.getSingleValue( args[0],
ec.getRowIndex(),
ec.getColumnIndex() ) ;
ValueEval v2 = OperandResolver.getSingleValue( args[1],
ec.getRowIndex(),
ec.getColumnIndex() ) ;
ValueEval v3 = OperandResolver.getSingleValue( args[2],
ec.getRowIndex(),
ec.getColumnIndex() ) ; principal = OperandResolver.coerceValueToDouble( v1 ) ;
rate = OperandResolver.coerceValueToDouble( v2 ) ;
years = OperandResolver.coerceValueToDouble( v3 ) ; result = calculateMortgagePayment( principal, rate, years ) ; checkValue(result); } catch (EvaluationException e) {
e.printStackTrace() ;
return e.getErrorEval();
} return new NumberEval( result ) ;
} public double calculateMortgagePayment( double p, double r, double y ) {
double i = r / 12 ;
double n = y * 12 ; //M = P [ i(1 + i)n ] / [ (1 + i)n - 1]
double principalAndInterest =
p * (( i * Math.pow((1 + i),n ) ) / ( Math.pow((1 + i),n) - 1)) ; return principalAndInterest ;
} /**
* Excel does not support infinities and NaNs, rather, it gives a #NUM! error in these cases
*
* @throws EvaluationException (#NUM!) if <tt>result</tt> is <tt>NaN</> or <tt>Infinity</tt>
*/
static final void checkValue(double result) throws EvaluationException {
if (Double.isNaN(result) || Double.isInfinite(result)) {
throw new EvaluationException(ErrorEval.NUM_ERROR);
}
} }

Great!  Now we need to go back to our original program that failed to evaluate our cell and add code that will allow it run our new Java code.

Registering Your Function

Now we need to register our function in the Workbook, so that the Formula Evaluator can resolve the name "calculatePayment" and map it to the actual implementation (CalculateMortgage). This is done using the UDFFinder object.  The UDFFinder manages FreeRefFunctions which are our analogy for the VBA code.  We need to create a UDFFinder. There are     a few things we need to know in order to do this:

  • The name of the function in the VBA code (in our case it is calculatePayment)
  • The Class name of our FreeRefFunction

UDFFinder is actually an interface, so we need to use an actual implementation of this interface.  Therefore we use the org.apache.poi.ss.formula.udf.DefaultUDFFinder class.  If you refer to the Javadocs you'll see that this class expects to get two arrays, one     containing the alias and the other containing an instance of the class that will represent that alias.  In our case our alias will be calculatePayment      and our class instance will be of the  CalculateMortgage type.  This class needs to be available at compile and runtime.  Be sure to keep these arrays     well organized because you'll run into problems if these arrays are of different sizes or the alias aren't in the same relative position in their respective     arrays.  Add the following code:

String[] functionNames = { "calculatePayment" } ;
FreeRefFunction[] functionImpls = { new CalculateMortgage() } ; UDFFinder udfs = new DefaultUDFFinder( functionNames, functionImpls ) ;
UDFFinder udfToolpack = new AggregatingUDFFinder( udfs ) ;

Now we have our UDFFinder instance and we've created the AggregatingUDFFinder instance.  The last step is to pass this to our Workbook:

workbook.addToolPack(udfToolpack);
     	  	

So now the whole class will look like this:

import java.io.File ;
import java.io.FileInputStream ;
import java.io.FileNotFoundException ;
import java.io.IOException ; import org.apache.poi.openxml4j.exceptions.InvalidFormatException ;
import org.apache.poi.ss.formula.functions.FreeRefFunction ;
import org.apache.poi.ss.formula.udf.AggregatingUDFFinder ;
import org.apache.poi.ss.formula.udf.DefaultUDFFinder ;
import org.apache.poi.ss.formula.udf.UDFFinder ;
import org.apache.poi.ss.usermodel.Cell ;
import org.apache.poi.ss.usermodel.CellValue ;
import org.apache.poi.ss.usermodel.Row ;
import org.apache.poi.ss.usermodel.Sheet ;
import org.apache.poi.ss.usermodel.Workbook ;
import org.apache.poi.ss.usermodel.WorkbookFactory ;
import org.apache.poi.ss.util.CellReference ; public class Evaluator { public static void main( String[] args ) { System.out.println( "fileName: " + args[0] ) ;
System.out.println( "cell: " + args[1] ) ; File workbookFile = new File( args[0] ) ; try {
FileInputStream fis = new FileInputStream(workbookFile);
Workbook workbook = WorkbookFactory.create(fis); String[] functionNames = { "calculatePayment" } ;
FreeRefFunction[] functionImpls = { new CalculateMortgage() } ; UDFFinder udfs = new DefaultUDFFinder( functionNames, functionImpls ) ;
UDFFinder udfToolpack = new AggregatingUDFFinder( udfs ) ; workbook.addToolPack(udfToolpack); FormulaEvaluator evaluator = workbook.getCreationHelper().createFormulaEvaluator(); CellReference cr = new CellReference( args[1] ) ;
String sheetName = cr.getSheetName() ;
Sheet sheet = workbook.getSheet( sheetName ) ;
int rowIdx = cr.getRow() ;
int colIdx = cr.getCol() ;
Row row = sheet.getRow( rowIdx ) ;
Cell cell = row.getCell( colIdx ) ; CellValue value = evaluator.evaluate( cell ) ; System.out.println("returns value: " + value ) ; } catch( FileNotFoundException e ) {
e.printStackTrace();
} catch( InvalidFormatException e ) {
e.printStackTrace();
} catch( IOException e ) {
e.printStackTrace();
}
}
}

Now that our evaluator is aware of the UDFFinder which in turn is aware of our FreeRefFunction, we're ready to re-run our example:

Evaluator mortgage-calculation.xls Sheet1!B4

which prints the following output in the console:

fileName: mortgage-calculation.xls
cell: Sheet1!B4
returns value: org.apache.poi.ss.usermodel.CellValue [790.7936267415464]

That is it!  Now you can create Java code and register it, allowing your POI based appliction to run spreadsheets that previously were inaccessible.

This example can be found in the src/examples/src/org/apache/poi/ss/examples/formula folder in the source.

[转]POI : How to Create and Use User Defined Functions的更多相关文章

  1. Read / Write Excel file in Java using Apache POI

    Read / Write Excel file in Java using Apache POI 2014-04-18 BY DINESH LEAVE A COMMENT About a year o ...

  2. [Hive - LanguageManual] Create/Drop/Alter -View、 Index 、 Function

    Create/Drop/Alter View Create View Drop View Alter View Properties Alter View As Select Version info ...

  3. PostgreSQL 之 CREATE FUNCTION

    官方文档 语法: CREATE [ OR REPLACE ] FUNCTION name ( [ [ argmode ] [ argname ] argtype [ { DEFAULT | = } d ...

  4. MySql批处理的小窍门:排行榜类数据生成

    MySql批处理的小窍门:排行榜类数据生成 最近在做新版本的开发,其中涉及到排行榜的批量预生成,在此分享给大家. 关键点 名次的计算(不考虑用游标) 单榜单查询 对于排行榜这种类型的数据,当只查一个排 ...

  5. 超级顽固的流方式读取doc,docx乱码问题

    因为工作中需要一个把doc或者docx的office文档内容,需要读取出来,并且也没展示功能.代码中第一考虑可能就是通过读取流方式,结果写了以后,各种乱码,百科的解决方案也是千奇百怪,第一点:可能是文 ...

  6. Phoenix综述(史上最全Phoenix中文文档)

    个人主页:http://www.linbingdong.com 简书地址:http://www.jianshu.com/users/6cb45a00b49c/latest_articles 网上关于P ...

  7. 【repost】JavaScript Scoping and Hoisting

    JavaScript Scoping and Hoisting Do you know what value will be alerted if the following is executed ...

  8. PHP7函数大全(4553个函数)

    转载来自: http://www.infocool.net/kb/PHP/201607/168683.html a 函数 说明 abs 绝对值 acos 反余弦 acosh 反双曲余弦 addcsla ...

  9. 有关binlog的那点事(mysql5.7.13)

    binlog作为mysql中最重要的日志之一,能实现异常恢复以及主从复制. 我们主要讨论的是主从复制中的binlog,这里将以mysql5.7.13的源码为主要依据来分析binlog. 在主从复制中, ...

随机推荐

  1. git bash的使用

    一.创建本地版本控制仓库 cd e:   进入e盘 cd gitspace 进入gitspace文件夹 git init 将E:\gitspace初始化为本地版本控制仓库 Initialized em ...

  2. python multiprocessing多进程应用

    multiprocessing包是Python中的多进程管理包,可以利用multiprocessing.Process对象来创建进程,Process对象拥有is_alive().join([timeo ...

  3. 转,如果linux不能用yum安装asterisk时,可以库参照以下办法添加asterisk仓库

    LinuxCentOSRedHat Installing a binary distribution of Asterisk makes it easier to maintain your syst ...

  4. JDBCTool

    新建 *.properties属性文件,内容如下: driver=com.mysql.jdbc.Driver url=jdbc:mysql://localhost:3306/dbName userna ...

  5. http 和 ajax 的关系

    经常听到有人混淆两者之间的基本概念,认为只是毫无关系的两种东西. 其实ajax也属于http请求,只不过是http一共有两种请求,一种是同步,一种是异步当然也可以同步那就是ajax. 总所周知hhtp ...

  6. AQS共享锁应用之Semaphore原理

    我们调用Semaphore方法时,其实是在间接调用其内部类或AQS方法执行的.Semaphore类结构与ReetrantLock类相似,内部类Sync继承自AQS,然后其子类FairSync和NoFa ...

  7. Python之线程与GIL

    前言            以下内容是个人学习之后的感悟,转载请注明出处~ 线程是什么 线程是程序中一个单一的顺序控制流程.进程内一个相对独立的.可调度的执行单元,是系统独立调度和分派CPU的 基本单 ...

  8. SQL 电子书

    http://vdisk.weibo.com/search/?type=&sortby=default&keyword=SQL+Server&filetype=&pag ...

  9. android实例3:拖动条

    个人网站http://www.ravedonut.com/ 拖动条改变图片的透明度 xml <LinearLayout xmlns:android="http://schemas.an ...

  10. 1.3-1.4 hive环境部署

    一. 官网:http://hive.apache.org/ 下载:http://archive.apache.org/dist/hive/ GitHub:https://github.com/apac ...